561988 - Can't install Jetpacks from a website

Status: RESOLVED INVALID

(Toolkit :: Add-ons Manager, defect)

Description

[Henrik Skupin [:whimboo][⌚️UTC+1]]

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.3a5pre) Gecko/20100426 Minefield/3.7a5pre

There is currently no way to install a jetpack from a website without having the jetpack engine installed separately. There is no notification bar shown when you open a page which contains a jetpack.

Steps:
1. Create a fresh profile
2. Open http://ehsanakhgari.org/blog/2010-01-12/bugzilla-made-even-more-awesome

A notification bar should be shown to let you install the jetpack.

Comment 1

[Dave Townsend [:mossop]]

It is not expected that you should be able to install jetpacks based on the old jetpack prototype in the new add-ons manager without the jetpack prototype installed.

Comment 2

[Henrik Skupin [:whimboo][⌚️UTC+1]]

That means with an updated jetpack (an example I will get from you) it already works? In that case it would be nice to get at least 2-3 differnet ones so we can run tests by installing multiple ones until an official page with jetpacks is available.

Comment 3

[Dave Townsend [:mossop]]

(In reply to comment #2)
> That means with an updated jetpack (an example I will get from you) it already
> works? In that case it would be nice to get at least 2-3 differnet ones so we
> can run tests by installing multiple ones until an official page with jetpacks
> is available.

The final specification for how they work only got decided a short time ago but yes I'm trying to get some made up. Since they aren't in the wild though I suspect it's not critical that we test much of these before the trunk landing.

Comment 4

[Henrik Skupin [:whimboo][⌚️UTC+1]]

(In reply to comment #3)
> The final specification for how they work only got decided a short time ago but
> yes I'm trying to get some made up. Since they aren't in the wild though I
> suspect it's not critical that we test much of these before the trunk landing.

If we detect a bug which requires a change of the specification, would that involve a schema change of the extension.sqlite? Or is that file immune to such a change?

Comment 5

[Dave Townsend [:mossop]]

(In reply to comment #4)
> (In reply to comment #3)
> > The final specification for how they work only got decided a short time ago but
> > yes I'm trying to get some made up. Since they aren't in the wild though I
> > suspect it's not critical that we test much of these before the trunk landing.
> 
> If we detect a bug which requires a change of the specification, would that
> involve a schema change of the extension.sqlite? Or is that file immune to such
> a change?

It is unlikely. But schema changes to extensions.sqlite aren't necessarily a problem once on trunk, we can write code to easily migrate between the schemas, I just haven't been doing that on the development branch because it is additional work and testing which seems mostly wasted when it is only a tiny group of people currently using the branch.